DESCRIPTION (Applicant's Abstract): This is a revision of 2 R01 MH47073-06, and is a proposal for continuation of work begun under a FIRST Award (5 R29 MH47073-05) to the P.I. The primary goal of this study is to test our theory concerning a disturbance in the processing of context in schizophrenia, and its relationship to prefrontal cortex (PFC) function. In previous work, we constructed neural network models of normal performance in a set of cognitive tasks in which schizophrenics were known to exhibit deficits. This resulted in a specific hypothesis concerning the function of PFC, and its involvement in schizophrenia: that PFC is responsible for the processing of context information, and that a disturbance in this mechanism is responsible for a number of the cognitive deficits observed in schizophrenia. Under the FIRST Award, we conducted behavioral experiments designed to test predictions made by this theory. In this study, and two others that we have recently completed, we observed a highly selective pattern of schizophrenic deficits in conditions sensitive to the processing of context, that provide strong preliminary support for our hypothesis. At the same time, our findings have raised a number of important new questions, that we address in the specific aims of this project. First, we propose a set of behavioral studies, that will test new predictions made by our current generation of computer simulation models. Second, we will test each subject three times, permitting a more sensitive assessment of the relationship between cognitive deficits and clinical symptoms. Third, we will test unmedicated, as well as medicated patients, to more carefully assess the influence that medications have on predicted deficits. Finally, we will conduct a neuroimaging experiment in a subset of subjects, using fMRI to study activation of PFC during performance of one of our cognitive tasks (AX-CPT). This will provide a more direct test of our hypothesis concerning the function of PFC, and its involvement in schizophrenia. Thus, this project will use a set of carefully constructed, theoretically motivated cognitive tasks to test specific predictions concerning the cognitive deficits in schizophrenia and, in a subcomponent, the relationship of PFC function to these deficits. We have already begun to demonstrate that our theory successfully predicts how tasks can be made more sensitive and specific to schizophrenic deficits. Positive results would represent a significant advance in our theoretical and empirical work, providing the groundwork for the development of powerful new behavioral and functional imaging tools for use in the diagnosis, assessment, and treatment of schizophrenia.